EGF Receptor Truncates

1. Field of the Invention

The Epidermal Growth Factor Receptor (hereinafter the EGF receptor) is known to be a 170,000 dalton glycoprotein (Reference No. 1) present on the surface of a variety of cell types. A gene for this receptor can be obtained by a process described in U.S. Pat. No. 4,933,294 (Reference No. 2) which is incorporated herein for this purpose. Like other known cell surface receptors, it has an extracellular and cytoplasmic (or intracellular) domain wherein the extracellular domain is capable of binding a substance so that the cytoplasmic domain interacts with another cellular molecule. That is, the interaction between the cytoplasmic domain of the EGF receptor with another cellular molecule is a function of the binding by the substance to the extracellular domain of the EGF receptor. The substance that binds the receptor is known as a ligand. Thus, the term "ligand" is limited only by an ability to affect the extra-cellular portion of the receptor so that the cytoplasmic domain is caused to interact with another molecule inside the cell. This interaction is termed "signal transduction".

For example, the signal transducing function of epidermal growth factor to the EGF receptor results in tyrosine phosphorylation and initiates a cascade of events that culminate in cell division essential in the growth and reproduction of the cell (Reference No. 3).

The cascade of events may be initiated by conformational changes of the receptor or receptor-receptor interactions effected by the ligand which modifies the affect of the receptor on cytoplasmic components.

The term ligand does not imply any particular molecular size or other structural or compositional feature. Furthermore, ligands may be natural or non-natural.

In an effort to find better techniques to design drugs, generally substances having low molecular weight, the research of the pharmaceutical industry is focusing on receptor technology and more specifically on ligands which bind to the receptor.

Conversely, the receptor truncates of the present invention are substances which bind ligands that offend an organism by binding the extracellular portion of an EGF receptor. Adsorptive capacities of the truncates may provide a benefit to an organism against detrimental invasion by an offending substance such as a virus or excess epidermal growth factor receptor ligand.

The present invention provides parts of the EGF receptor to which ligands bind for ligands that bind the EGF receptor itself. The parts are a glycoprotein having selected EGF receptor sequences preferably having substantially fewer amino acids than the EGF receptor itself such that the glycoprotein is relative low molecular weight. In other words, the glycoproteins of the present invention bind ligands of the EGF receptor, but are preferably smaller than an EGF receptor and are soluble proteins and as so may bind a natural ligand without transmitting a signal for the growth and reproduction of a cell.

Among the known natural ligands which bind specifically to the EGF receptor and initiate a cellular growth response are: Epidermal Growth Factor, Transforming Growth Factor .alpha. (Reference No. 4) Amphiregulin (Reference No. 5), and Vaccinia Growth Factor (Reference No. 6.)

Thus, the present invention is portions of the extracellular domain of the EGF receptor which are produced using recombinant DNA methods to yield soluble proteins capable of binding the natural ligands.

2. Description of the Prior Art

The extracellular domain of the EGF receptor is described as an external EGF binding domain comprising 621 amino acids (Reference No. 7). On the basis of amino acid sequence analysis, this extracellular region of EGF receptor can be divided into four domains. It has been suggested that the extracellular portion of the EGF receptor which is flanked by two cysteine-rich domains contributes most of the interactions that define ligand-binding specificity of the EGF receptor (Reference No. 8). The four domains are identified as D.sub.1, D.sub.2, D.sub.3, and D.sub.4.

The portion of the EGF receptor defining ligand binding for the receptor is said to be the D.sub.3 domain within the EGF receptor.

The present invention provides an expressed protein which is the D.sub.3 domain. This protein and additional proteins of the present invention define an independent portion of the EGF receptor which is a properly folded protein capable of ligand binding. Such a protein, that is only a portion of the whole EGF receptor, is not previously described.

The present invention may use either an EGF receptor cDNA sequence (Reference No. 1) or the EGF receptor gene (Reference No. 9) and their expression which are obtained and carried out respectively by conventional methods.

The present invention is a novel protein having selected amino acid sequences of the EGF receptor. These sequences are meant to include functional equivalents of amino acids present in the sequences.

Additionally, the present invention embodies any DNA sequence coding for the novel proteins presented in FIGS. 4, 7, and 10.

These may be `mini-genes`. The `mini-genes` are cDNA sequences coding for the desired protein which is flanked by a promotor on the 5' side of the sequence and a polyadenylation region on the 3' side of the cDNA. The cDNA sequences are as disclosed or are also meant to extend to functional equivalents thereof.

These proteins may also be prepared through the use of the genomic DNA sequence encoding the EGF receptor.

Thus, the present invention is also a process for the preparation of the selected amino acid sequences of the EGF receptor.

The amino acid sequences are set out in accordance with the three-letter abbreviations of 37 CFR 1.822(b)(2) and the truncate codes are set out in accordance with the base codes of 37 CFR 1.822(b)(1).

FIG. 1 represents the construction of plasmid RC-2b having NRRL No. B-18718 deposited Sep. 26, 1990, the mammalian expression vector used for the introduction of EGF receptor truncate `mini-genes` into cells and is described in the Experimental Section hereinafter.

FIG. 2 represents the plasmid map of pRLD.sub.1 D.sub.2 D.sub.3.ApaL which notes the location of the EGF leader, D.sub.1, D.sub.2, and D.sub.3 domains of the EGF receptor gene and ApaL region inserted into the RC-2b expression plasmid as described in the Experimental Section.

FIG. 3 shows the DNA sequence of the pRLD.sub.1 D.sub.2 D.sub.3.ApaL construct as described in the Experimental Section.

FIG. 4 presents the amino acid sequence of the LD.sub.1 D.sub.2 D.sub.3.ApaL protein, deduced from the DNA sequence shown in FIG. 3.

FIG. 5 represents the plasmid map of pRLD.sub.2 D.sub.3 D.sub.4 which notes the location of the EGF leader, D.sub.2, D.sub.3, and D.sub.4. domains of EGF receptor gene.

FIG. 6 shows the DNA sequence of the pRLD.sub.2 D.sub.3 D.sub.4 construct as described in the Experimental Section.

FIG. 7 presents the amino acid sequence of the LD.sub.2 D.sub.3 D.sub.4 protein, deduced from the DNA sequence shown in FIG. 6.

FIG. 8 represents the plasmid map pLRD.sub.3 D.sub.4 which notes the location of the EGF leader, D.sub.3 and D.sub.4 domains of the EGF receptor gene, as described in the Experimental Section.

FIG. 9 shows the DNA sequence of the pRLD.sub.3 D.sub.4 construct as described in the Experimental Section.

FIG. 10 presents the amino acid sequence of the LD.sub.3 D.sub.4 protein deduced from the DNA sequence shown in FIG. 9.

FIG. 11 (a and b) shows a radiograph of SDS-Polyacrylamide gel electrophoresis analysis of immunoprecipitates from the .sup.35 S labeled supernatants of Cos cells electroporated with the EGF receptor truncate genes as described in the Experimental Section. FIG. 11a depicts the immunoprecipitation with Ab-1 and FIG. 11b depicts the immunoprecipitation with RPN.513.

FIG. 12 shows an experiment similar to that described for FIG. 11 using an EGF-affi gel matrix as described in the Experimental Section.

A glycoprotein having selected EGF receptor sequences but which is capable of binding ligands of an EGF receptor are produced in mammalian cells by recombinant DNA methodology. Selection of desired sequences may be limited only by the requirement that ligands which bind EGF receptor must also bind the desired sequences. Preferably, the sequences are a small single region of the receptor of about a 20kd polypeptide.

Additionally, although mammalian cells provide a host for the production of glycoproteins, the glycosylation sites which influence folding of such glycoproteins may be unnecessary for reasonable affinity (100 times less or better in EGF binding) compared to the EGF receptor. Thus, recombinant proteins produced by mammalian cell culture, yeast culture, bacterial culture, and the like recombinant DNA technology also may provide recombinant receptor proteins which can bind EGF or similar ligands.

`Mini-genes` coding for receptor truncates are readily adapted to high-level stable expression in Chinese Hamster ovary cells by insertion of a Dihydrofolate Reductase Gene Cassette into the vector having such mini-genes and by methotrexate induced gene amplification of transfected cells. In addition, any known or commercially available mammalian expression system including readily available plasmids, can be used. Alternatively, the DNA sequence coding (described herein) for the receptor truncates may be excised from the mammalian expression vector and placed into vectors to provide high level expression in yeast, bacterial or insect cell systems.

The mini-genes are preferably designed to produce secreted recombinant proteins from cells. Thus, it is preferred to construct `mini-genes` coding for the desired protein and express these truncates transiently in Cos cells for analysis. Such proteins may be metabolically labeled with .sup.35 S-methionine and cysteine followed by analysis of truncate proteins present in the culture media by immunoprecipitation with various anti-sera or monoclonal antibodies.

Ligand binding is assessed by an EGF or .alpha.TGF affinity matrix.

The present invention is also a process for selection of desired sequences for the production of proteins which are portions of the EGF receptor having the capability of binding ligands by further truncation.

This selection process is designed to provide increasingly smaller portions of the EGF receptor which retain the binding capability. For example, further weak reactivity with anti-sera raised against native EGF receptors or monoclonal antibodies RPN.513 or Ab-1 indicated a small amount of properly folded receptor protein was obtained from LD.sub.3.APAL and LD.sub.1 D.sub.2 D.sub.3 constructs. Truncates containing fewer sequences than those required to express the D.sub.3 protein in these documented truncates were found not to be expressed at high enough levels in mammalian cells to examine EGF binding. In like systems expressing like proteins, this indicates improper folding and rapid degradation in the cells. Alternate expression strategies can correct this problem. Thus, the present invention also includes LD.sub.3.APAL.

Additionally, two truncates which code for the smaller portions of the EGF receptor are found to bind EGF (LD.sub.1 D.sub.2 D.sub.3.APAL and LD.sub.3 D.sub.4 described hereinafter) and these proteins have an overlapping sequence. However, it is found that expression of the overlapping sequence LD.sub.3.APAL is greatly diminished, presumably because there is insufficient amino acid sequence necessary to adequately fold the D.sub.3 EGF binding domain in Cos cells. The process of the present invention encompasses a previously unpredictable structural profile of the EGF receptor.

The following examples are offered by way of illustration and not by way of limitation.

The construction of `mini-genes` coding for three novel recombinant glycoproteins containing EGF receptor sequences and preliminary characterization of their protein products is described.

All DNA manipulations were done according to standard procedures. See Molecular Cloning, T. Maniatis et al., Cold Spring Harbor Lab., 1982. Enzymes used in cloning were obtained either from New England Biolabs or Bethesda Research Laboratories and employed according to the supplier's directions.

1. LD.sub.1 D.sub.2 D.sub.3. ApaL. This EGF receptor truncate codes for a protein encompassing amino acids Met.sub.-24 to Val.sub.505 of the EGF receptor. Amino acid +1 designates the mature amino terminus of the receptor. The first step of this multiple step construction strategy is to insert a 1567 base pair (bp) Sacl-Apal fragment of the EGF receptor cDNA into Sacl-Eco R1 cleaved pSP-65 (Promega) with synthetic oligonucleotide linkers (shown as capital letters) as shown. ##STR1##

This step introduces a stop codon after amino acid 494 of the receptor and creates a Hind 3 site 5' of the receptor coding sequence (from the pSP.6S polylinker) and a Bst X1 site 3' of the coding sequence. This construct is designated pSLD.sub.1 D.sub.2 D.sub.3. The 1623 bp. Hind 3-Bst X1 EGF receptor fragment of pSLD.sub.1 D.sub.2 D.sub.3 is inserted into Hind 3+Bst X1 cleaved RC-2b mammalian expression vector (FIG. 1). This resultant plasmid, pRLD.sub.1 D.sub.2 D.sub.3 is further modified by inserting the indicated oligonucleotide linkers into Apal+ Bst X1 cut pRLD.sub.1 D.sub.2 D.sub.3 resulting in the construction of pRLD.sub.1 D.sub.2 D.sub.3.Apa L. Restriction site mapping and DNA sequencing confirms that this construct is as designed. The plasmid map and DNA sequence of pRLD.sub.1 D.sub.2 D.sub.3.Apa L is presented in FIGS. 2 and 3. The amino acid sequence of the LD.sub.1 D.sub.2 D.sub.3.Apa L protein, deduced from DNA sequence, is presented in FIG. 4. ##STR2##

2. LD.sub.2 D.sub.3 D.sub.4. This EGF receptor truncate is designed to meld the EGF receptor leader sequence to the receptor protein beginning at Ser.sub.150. This protein possesses the leader peptide (amino acids Met.sub.-24 to Ala.sub.-1), 9 amino acids of the mature amino terminus of the receptor (Leu.sub.1 to Gly.sub.9) and receptor sequences Ser.sub.150 to Gly.sub.625. In order to construct the gene for this protein a 1961 bp. Sacl-Bst X1 fragment encoding the entire EGF receptor extracellular domain is inserted into Sacl-Eco R1 cleaved pSP-65 with the oligonucleotide linkers as shown, resulting in the plasmid pSLD.sub.1 D.sub.2 D.sub.3 D.sub.4. ##STR3##

A 1423 bp Af13-Bst X1 fragment of pSLD.sub.1 D.sub.2 D.sub.3 D.sub.4 (coding for amino acids 150 to 625 of the receptor) is ligated to a 3107 bp Sty 1-Bst X1 fragment of pSLD.sub.1 D.sub.2 D.sub.3 (contains DNA of pSP-65 vector, EGF receptor leader sequence and Leu.sub.1 to Gly.sub.9)with the linkers as shown, resulting in pSLD.sub.2 D.sub.3 D.sub.4. ##STR4##

Transfer of this truncate gene from pSP-65 to RC-2b is accomplished by inserting the 1596 bp Hind3-Bst X1 fragment of pSLD.sub.2 D.sub.3 D.sub.4 into Hind3+Bst X1 cleaved RC-2b resulting in pRLD.sub.2 D.sub.3 D.sub.4. Restriction site mapping and DNA sequencing confirms that this construct is as designed. The plasmid map and DNA sequence for pRLD.sub.2 D.sub.3 D.sub.4 is presented in FIGS. 5 and 6. The amino acid sequence of LD.sub.2 D.sub.3 D.sub.4 deduced from the DNA sequence, is presented in FIG. 7.

3. LD.sub.3 D.sub.4. This EGF receptor protein possesses the leader peptide (Met.sub.-24 to Ala.sub.-1), Leu.sub.1 to Gln.sub.8 of the mature amino terminus of the receptor and receptor sequence Asp.sub.297 to Gly.sub.625. To construct the gene for this protein a 587 bp Aha2-Bst X1 fragment of pRLD.sub.1 D.sub.2 D.sub.3 (coding for amino acids Asp.sub.297 to Pro.sub.494) is ligated to the 3107 bp Styl-Bst X1 fragment of pSLD.sub.1 D.sub.2 D.sub.3 with the oloigonucleotide linkers as shown. ##STR5##

This plasmid, pSLD.sub.3, is subsequently cleaved with Hind3+Bst X1 and the 758 bp fragment (encoding amino acids -24 to 494) is ligated to Hind3+Bst X1 cut RC-2b. This plasmid is designated pRLD.sub.3. pRLD.sub.3 D.sub.4 is constructed by inserting a 406 bp Apal-Bst X1 fragment from pRLD.sub.2 D.sub.3 D.sub.4 into the 5357 bp Apal-BstX1 fragment of pRLD.sub.3. Restriction site mapping and DNA sequencing confirms that this construct is as designed. The plasmid map and DNA sequence for pRLD.sub.3 D.sub.4 is presented in FIGS. 8 and 9. The amino acid sequence of the LD.sub.3 D.sub.4 protein, deduced from the DNA sequence is presented in FIG. 10.

Transient expression of these EGF receptor truncate proteins is accomplished by electroporation (Reference No. 11) of Cos 1 cells (Reference No. 12) at 3.times.10.sup.6 cells/ ml. with 20 ug. of CsCl purified (Reference No. 13) plasmid DNA using a BRL Cell-porator set at 330 uF, 300 volts at low Ohms. Following a 2 minute incubation at room temperature, the electroporated cells (0.3 ml.) are seeded in a 100 mm tissue culture plate containing 10 ml. of DMEM+10% Fetal Calf Serum. Forty eight hours after the electroporation the media is changed to labelling media consisting of RPMI (without Methioine and cysteine), 5% dialyzed fetal calf serum, 50 uCi/ml, each of .sup.35 S- Methioine and .sup.35 S- Cysteine and 50 ug/ml. gentamycin. After a 24 hour labelling period, the media is removed, Phenyl Methyl Sulphonyl Fluoride (a protease inhibitor) is added to 0.2 uM. and centrifuged at 10,000 rpm for 10 minutes at 4.degree. C. The supernatant is removed and used for immunoprecipitation studies (Reference No. 13) with EGF receptor specific monoclonal antibodies Ab-1(Oncogene Science) and RPN.513 (Amersham Corp.). EGF binding abilities of these proteins is assessed by incubation with an EGF-affinity matrix prepared by cross-linking EGF with Affi-gel 10 (Biorad Laboratories). FIG. 11 (a and b) shows a radiograph of SDS-Polyacrylamide gel electrophoresis analysis of immunoprecipitates from the .sup.35 S labelled supernatants of Cos cells electroporated with the EGF receptor truncate genes. FIG. 11a depicts immunoprecipitation with Ab-1 and FIG. 11b uses RPN.513. This figure shows that a 68,000 dalton protein from cells electroporated with pRLD.sub.1 D.sub.2 D.sub.3.Apa L is specifically precipitated with either of the monoclonal antibodies whereas 85,000 and 55,000 dalton proteins from cells electroporated with pRLD.sub.2 D.sub.3 D.sub.4 or pRLD.sub.3 D.sub.4 respectively are specifically precipitated with only the Ab-1 monoclonal. FIG. 12 shows a similar experiment using the EGF-affi gel matrix. This experiment indicates that the 3 receptor truncate proteins, LD.sub.1 D.sub.2 D.sub.3.Apa L, LD.sub.2 D.sub.3 D.sub.4 and LD.sub.3 D.sub.4 bind the EGF matrix while cells electroporated with pRLD.sub.1 D.sub.2 D.sub.3, LD.sub.3.ApaL or RC-2b do not produce a protein capable of binding EGF.

The recombinant proteins described in the preceding section can be over-expressed in any of a number of heterologous cell systems including a Chinese Hamster Ovary system using Dihydrofolate Reductase co-amplification (Reference No. 14) or a Baculovirus - insect cell system (Reference No. 15).

The purified proteins of the present invention, such as LD.sub.1 D.sub.2 D.sub.3.Apa L, LD.sub.2 D.sub.3 D.sub.4 and LD.sub.3 D.sub.4 can be used as an adsorptive agent for any moieties that bind the EGF receptor, such as EGF, TGF-.alpha., and any as yet undiscovered peptides or etiological agents, such as a virus that uses the EGF receptor as the portal of entry to the cell. As an adsorptive agent these proteins will compete with the EGF receptor present on the cell surface for binding of the ligands and thereby inhibit the action of the ligands.

For example, it is accepted that epidermal growth factor binding induces a conformation change in the external domain of its receptor (Reference No. 9); thus, excess epidermal growth factor bound to the proteins of the present invention would prevent effects on an organism's endogenous EGF receptor by EGF binding. In like manner, recognition of EGF receptor's participation as a portal for infectivity for vaccinia virus is disclosed by N. H. Colburn et al. (Reference No. 16) so addition of the present proteins providing absorption of the virus may prevent entry into otherwise susceptible cells (Reference No. 17).

Of course, the EGF receptor truncate proteins of the present invention are also useful in any manner previously set out for use of EGF receptor itself. For example, the use described for the EGF receptor itself in a method for the detection of abnormalities in mammalian cell growth is described in U.S. Pat. No. 4,933,294 (Reference No. 2) which is therefor incorporated herein by reference.

Also, EGF receptor truncate proteins of the present invention are useful to prepare novel receptors for efficient determination of ligands and their antagonists or agonists as described in U.S. Pat. No. 4,859,609 (Reference No. 17) which is also incorporated by reference, therefor.

1. A. Ullrich, L. Coussens, J. S. Hayflick, T. J. Dull, A. Gray, A. W. Tam, J. Lee, Y. Yarden, T. A. Libermann, J. Schlessinger, J. Downward, E. L. V. Mayes, N. Whittle, M. D. Waterfield, and P. H. Seeburg; "Human epidermal growth factor receptor cDNA sequence and aberrant expression of the amplified gene in A431 epidermoid carcinoma cells." Nature, Vol. 309, 418-425 (31 May 1984).

2. U.S. Pat. No. 4,933,294, M. D. Waterfield, et al. issued Jun. 12, 1990.

3. EMBO J., September 1986 5(9), p 2179-90.

4. Todaro, G. J.; Fryling, C.; and Delarco, J. E.; "Transforming growth factors produced by certain human tumor cells: polypeptides that interact with epidermal growth factor receptors." PNAS 198077: p 5258-5262

5. Shayab, M.; Plauman, G. D.; McDonald, V. L.; Bradley, J. G.; Todaro, G. J.; "Structure and Function of Human Amphiregulin: A Member of the Epidermal Growth Factor Family" Science 1989, 243, p 1074.

6. Stroobant, P.; Rice, A. P.; Gullick, W. J.; Cheng, D. J.; Kerr, I. M.; Waterfield, M. D.; "Purification and Characterization of Vaccinia Virus Growth Factor." Cell 1985, 42, p 383-393.

7. Proc. R. Soc. Lond [Biol.] Oct. 22, 1985, 226 (1242) p 127-34.

8. Mol All Biol. April 1988 8(4) p 1831-4.

9. EMBO J., 1989, August 13, p 4115-4123).

10. Toneguzzo, F.; Hayclay, A. D.; Keating, A.; 1986 Mol Cell Biol 6. p 703

11. Gluzman, Y., "SV-40 transformed simian cells support the replication of early SV-40 mutants." Cell 1981 23, p 175-182.

12. Maniatis, T.; Fritsch, E. F.; Sambrook, J.; Molecular Cloning: A laboratory manual. 1989 Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

13. Wigler, M.; Perucho, M.; Kurtz, D.; Dana, S.; Pellicer, R.; Axel, R.; Silverstein, S. "Transformation of mammalian cells with an amplifiable dominant acting gene." PNAS 1980, 77, p 3567-3570.

14. U.S. Pat. Nos. 4,745,051 and 4,879,236.

15. Inst. Bio-Organic Chem., Syntex, Res. Palo Alto, Calif. 94304, Proceedings of a Triton BioSciences UCLA Sym., Steamboat Springs, Colo., April 6-13, 1986, XXI, +388p.

16. Nature, (LOND) 318 (6047) 1985. p 663-5.

17. U.S. Pat. No. 4,859,609, issued Aug. 22, 1989 to J. Dull, et al.